Do automated accessibility checkers find every problem?
No. They catch perhaps a third of issues, and will not tell you whether a form works with a screen reader or whether the main task can be completed with a keyboard. Someone has to try.
Blog
Why accessibility is a governance issue, not a design extra
Because for most organisations it is a legal duty that nobody has been made responsible for. Equality legislation requires reasonable adjustments for disabled people, and that applies to digital services as it does to physical ones. When a website is challenged, "the agency built it" does not protect a board. Accessibility failures are usually structural, so building it in costs little and fixing it later is a project. A board needs to know which standard it works to, who confirmed it, when it was last tested and what remains unfixed.
Why does accessibility belong at board level?
Because it is a duty, nobody owns it, and the exposure sits with the board. Accessibility usually appears in a proposal as a line item near the bottom, often optional, sometimes priced separately, which tells you it is being treated as a design preference. For most of the organisations I work with it is a legal duty. I am not a lawyer and none of this is legal advice, but you do not need to be one to notice that a board has accepted a risk without discussing it.
In the UK, service providers have obligations under equality legislation to make reasonable adjustments for disabled people, and those obligations apply to digital services as they do to physical ones. Public sector bodies carry additional and more specific requirements. Whether a particular organisation is caught, and how, depends on what it is and what it does, which is why it needs a proper conversation rather than an assumption.
Ask who signed off that your website meets the required standard and you will usually get a pause. The agency assumed the client would specify it; the client assumed the agency would handle it. The result is a service nobody has assessed, with no record of a decision either way. If it is challenged, "the agency built it" is an uncomfortable position, and for trustees and non-executive directors with personal accountability for oversight it is more uncomfortable still.
Why does leaving it late cost so much?
Because accessibility failures are structural rather than cosmetic. The pattern I see most often: an organisation commissions a website, accessibility is missing from the brief or appears as a general aspiration, and the agency builds something modern and broadly fine. Nobody tests it with a screen reader, checks colour contrast or tries to complete the main task using only a keyboard.
Two years later somebody asks the question, and the fix costs many times what building it correctly would have. Contrast is a colour choice and cheap to change. Heading hierarchy, form labelling, focus order and how a component behaves for somebody not using a mouse are baked into the build. That is the argument to make to a finance director: building it in costs very little, and retrofitting it is a project.
Why is this sharper for charities and housing associations?
Because the organisations most likely to have accessibility problems are often the ones serving the people most likely to be affected. That is a resource problem rather than hypocrisy: smaller organisations buy from smaller suppliers on tighter budgets, and what gets cut is whatever nobody in the room is advocating for. It is still worth naming, because a gap between mission and delivery reads badly when it surfaces, and it usually surfaces publicly.
For a charity board the question is simple. If we exist to serve people including disabled people, and our main digital service excludes some of them, is that on the risk register, and who is accountable?
What should a board ask about accessibility?
Four questions, none of them technical.
Which standard are we working to, and who confirmed we meet it? The common reference point is WCAG 2.2 at level AA. The useful half of the question is the second: somebody's name should be attached to the assessment.
When was it last tested, and how? Automated checking tools catch perhaps a third of issues. They will not tell you whether a form is usable with a screen reader or whether the main task can be completed with a keyboard, so ask whether anybody has actually tried.
What did we find, and what have we not fixed? An organisation with a list of known issues and a plan is in a defensible position. One with no list is not, because it cannot show it ever looked.
Is it in the specification for the next thing we build? This is where the money is saved. Specified up front, it is a constraint the supplier designs around. Added later, it is a change request.
What should a board avoid?
A compliance exercise that produces a document and changes nothing. That is a real risk once a board takes an interest, and it is worse than doing nothing because it creates the appearance of control.
Treating accessibility as a cost centre is the other mistake. Most of what makes a site accessible makes it better generally: clear heading structure helps every reader and search engines, sensible contrast helps anybody on a phone in daylight, and forms that work with a keyboard work faster for everyone. Sites that are genuinely usable tend to convert better, and I have never seen the reverse.
I raise accessibility in reviews as a governance finding rather than a technical one, because that is the level at which it gets fixed: given to a marketing team it becomes a backlog item, given to a board with a name attached it gets resolved. I have no software to sell and no development team to keep occupied, and often the honest recommendation is a handful of specific fixes that move an organisation from exposed to defensible, at a fraction of the cost of a rebuild. A digital health analysis includes an accessibility assessment alongside performance and technical review, with the findings framed for a board.
Related evidence: DementiaNet: building a dementia support platform people actually use. Earlier in this journey: Digital on a charity budget.
Questions people ask
Does accessible design make a website better for everyone?
Generally, yes. Clear heading structure helps every reader and search engines, sensible contrast helps anyone on a phone in daylight, and forms that work with a keyboard work faster for everyone.
Not sure where to start?
Most clients begin with a conversation. No pitch, no hard sell.
Just a straightforward discussion about where you are and whether I can help.
Book a free 30-minute call